关键词: CO₂通量, 净初级生产力, 净生态系统生产力, 涡动协方差, 不确定性

Abstract:

Objective: Long-term forest CO₂ fluxes measured by eddy covariance(EC) were rarely validated with ground monitoring of carbon processes, which limited the understanding for the estimate accuracy of temperate forest carbon sink. This study mainly determined the long-term CO₂ fluxes and ground carbon processes and their uncertainties in a temperate secondary forest in northeast China. Method: Based on the CO₂ flux measurements of the typical temperate secondary forest at the Maoershan site over 11 years (2008—2018) via the EC method, we calculated the CO₂ fluxes and its uncertainty with site-specific approaches of data analysis; Based on 12 years’ inventory measurements during 2008—2019, we systematically estimated the net primary productivity(NPP) and net ecosystem production(NEP) and their uncertainties, and compared the CO₂ fluxes estimates between the two methods. Result: 1) Based on the EC method, the mean net ecosystem exchange(NEE), gross primary production(GPP), and ecosystem respiration(R_e) of the stand over the 11 years were ?1.57 ± 0.64, 13.56 ± 1.48, and 12.00 ± 1.38 t?hm^?2 a^?1, respectively; of which the corresponding uncertainties were 0.47, 0.90 and 1.37 t?hm^?2 a^?1, and the relative uncertainties were 29.9%, 6.6% and 11.4%. 2) Based on the inventory method, the total NPP of ecosystem was 7.54 ± 1.31 t?hm^?2 a^?1. The NPP of total litterfall and total woody tissues were 2.32 ± 0.14 t?hm^?2 a^?1 and 2.36 ± 0.14 t?hm^?2 a^?1, respectively, accounted for 30.8% and 31.3% of the total NPP, respectively. The understory NPP was 0.39 ± 0.04 t? hm^?2 a^?1, only accounted for 5.2% of the total; of which the NPP of shrubs and herbs were 0.08 ± 0.02 t?hm^?2 a^?1 and 0.31 ± 0.03 t?hm^?2 a^?1, respectively. The fine root NPP was 2.47 ± 1.29 t? hm^?2 a^?1, which was the largest fraction (32.7%) of total NPP. 3) For the canopy NPP, the uncertainty was mainly from litterfall production. Measurement of diameter at breast height, temporal and spatial variation of growth and carbon concentration accounted for 5.9%, and 32.3%, 65.2% and 2.5% of the uncertainty of wood tissue NPP for one-year measurement, respectively. The relative uncertainty of shrub, herb, and fine root NPP were 25.0%, 12.7% and 52.3%, respectively. 4) Combined with the soil heterotrophic respiration (5.23 ± 0.20 t?hm^?2 a^?1) and coarse woody debris respiration (0.37 ± 0.21 t?hm^?2 a^?1), the NEP was estimated to be 1.94 ± 1.34 t?hm^?2 a^?1, overlapped with the interval of absolute value of NEE estimate (1.57 ± 0.47 t?hm^?2 a^?1). Conclusion: The 11-year mean NEE measured by the EC method is closed to the NEP measured by the inventory method in the typical temperate secondary forest at the Maoershan site. Careful system design and site-adapted data analysis of the EC method, measuring all components of NPP as completely as possible via appropriate approaches, and extending the period length of measurements, will help to improve the accuracy of carbon flux measurement in forest ecosystems. These findings improve the understanding of the CO₂ fluxes in typical temperate deciduous forests in Chin and the uncertainty of forest NPP and NEP at the local scale, and provide data and methodological support for accurate assessment of ecosystem carbon cycling.

Key words: CO₂ fluxes, net primary productivity, net ecosystem productivity, eddy covariance, uncertainty